一种用于动态视频超分辨率的多尺度最小二乘仿射块匹配图像配准方法
doi: 10.3724/SP.J.1146.2007.01745
A Least Square Affine-Based Block-Match Image Registration Algorithm for Dynamic Video Super Resolution
-
摘要: 动态视频的超分辨率复原中,连续各帧图像间的精确匹配具有非常重要的意义。该文提出一种基于多尺度最小二乘仿射块匹配的图像配准方法。首先定义了一个指标Dmv来衡量图像的整体和局部匹配效果,并以此为基础设计了一种多尺度块选择机制,根据图像的运动情况选择匹配块大小,以兼顾图像中运动平坦和非平坦区域的匹配效果。与传统的块匹配方法不同,该文采用基于仿射模型的最小二乘配准方法实现各图像块的匹配,并通过修正步长的归一化处理解决了不同大小图像块在匹配时的收敛问题,从而在提高参数估计精度的同时降低了算法的运算量。最后,通过实验对算法的匹配性能及其对超分辨率复原算法整体性能的影响进行了测试。实验结果表明,该方法不仅可以实现更为准确的运动估计,当用于最大后验概率MAP超分辨率复原算法时,能够进一步有效提高算法的复原性能和实现速度。Abstract: Registration of the consecutive frames is quite essential in dynamic video super resolution. In this paper, a multi-scale least square affine-based block-match method is proposed. An index Dmv is defined to evaluate global and local matching performances of the images. Then a multi-scale scheme is designed to adjust the block size automatically according to the motions between frames, which guaranteed well performances in both planar and un-planar regions. Different from the traditional block-match method, the affine-based least square estimation algorithm is introduced for registration of each pair of blocks. Convergence of the estimating process for different size of blocks is resolved by unification of the update step, which results in improvements in both estimating precision and speed. Finally, the proposed algorithm is evaluated in both the registration performance and its affects to the performance of the super resolution algorithm. Experimental results show that the proposed algorithm not only can provide more accurate motion estimation, when be applied to the Maximum A Posterior (MAP) based super resolution method, it shows obvious enhancement in both reconstruction performances and efficiencies.
-
Park S C, Park M K, and Kang M G. Super-resolution imagereconstruction: A technical review [J].IEEE SignalProcessing Magazine.2003, 20(3):21-36[2]Hardie Russell C, Barnard Kenneth J, and Bognar John G,et al.. High resolution image reconstruction from a sequenceof rotated and translated frames and its application to aninfrared imaging system [J].Optical Engineering.1998, 37(1):247-260[3]Blerling M and Thema R. Motion compensating fieldinterpolation using a hierarchically structured displacementestimator [J].Signal Processing.1986, 11(4):387-404[4]Patanavijit V and Jitapunkul S. An iterative superresolutionreconstruction of image sequences using abayesian approach with BTV prior and affine block-basedregistration [C]. The 3rd Canadian Conference on Computerand Robot Vision(CRV06), Quebec City, June 7-9, 2006:45-51.[5]沈兰荪, 卓力. 小波编码与网络视频传输[M]. 北京: 科学出版社, 2005: 146-164.Shen Lan-sun and Zhuo Li. Wavelet Coding and VideoTransmission[M]. Beijing: Science Press, 2005: 146-164.[6]Barreto D, Callico G M, and Lopez S, et al.. Real-timesuper-resolution over raw video sequences [J]. Proceedings ofSPIE, Vol.5837 VLSI Circuits and Systems II, 2005: 628-637.李弼程, 彭天强, 彭波等. 智能图像处理技术[M]. 北京: 电子工业出版社, 2005: 260-290.Li Bi-cheng, Peng Tian-qiang, and Peng Bo, et al.. IntelligentImage Processing[M]. Beijing: Publishing House ofElectronics Industry, 2005: 260-290.[7]Schultz R R and Stevenson R L. Extraction of highresolutionframes from video sequences [J].IEEE Trans. onImage Processing.1996, 5(6):996-1011[8]张晓玲, 沈兰荪, Lam Kin-Man. 一种基于分形码和模型约束的图像放大算法[J]. 电子学报, 2006, 34(3): 433-436.Zhang Xiao-ling, Shen Lan-sun, and Lam Kin-Man. Imagemagnification based on fractal codes and model constraint [J].Acta Electronica Sinica, 2006, 34(3): 433-436.
计量
- 文章访问数: 4010
- HTML全文浏览量: 105
- PDF下载量: 2173
- 被引次数: 0